MOSS-VL-Realtime is Now Open Source
victormustar · x · 2026-07-14
MosiAI has open-sourced MOSS-VL-Realtime, an 11B vision-language model designed for real-time visual understanding of continuous video streams under the Apache-2.0 license.
It supports asking questions at any point during the video and features a "watch and answer" capability: it can correct or interrupt its answers when the scene changes, and remain silent when evidence is insufficient. The project also offers Base, Instruct, and Realtime versions, supporting bilingual (Chinese and English) multimodal understanding with a context window length of 256K tokens.
Related event: MOSS-VL-Realtime Open-Sourced for Real-time Vision(2 posts)→
More from Multimodal
- 3D ResNet Paper Crosses 3,000 Citations Eight Years After CVPR 2018 — HirokatuKataoka · 2026-09-11
- Non-coder builds full-featured Android ComfyUI client with ChatGPT, submits to Google Play — ComfierUI · 2026-09-11
- RunningHub open-sources H3Lightning, speeding up MiniMax H3 video generation 12x — 智东西 · 2026-09-11
- FastH3-Live hits 22fps: acceleration node benchmarks and the --vram-headroom trick — spartong945 · 2026-09-11
- Midjourney style code share: --sref 2912175708 — tisch_eins · 2026-09-11
- Astra storyboards plus Minimax H3 per-shot generation boost video success rates — Hailuo_AI · 2026-09-11